Апликабилност на антимилериановиот хормон како маркер во дијагнозата на полицистичниот оваријален синдром и неговата асоцијација со индексите на инсулинска резистенција

Abstract

Research focus: The study researches the role of anti-müllerian hormone (AMH) in the impaired ovarian steroidogenesis in women with polycystic ovary syndrome (PCOS) and the disfunctions in the insulin action that have impact on the metabolic abnormalities in PCOS. Objective: To evaluate the diagnostic potential of the anti-müllerian hormone аs а marker in the diagnosis of the polycystic ovary syndrome, to evaluate the association between AMH, androgens and indexes of insulin resistance in PCOS and to evaluate the diagnostic potential of HOMA-IR in determining insulin resistance in PCOS. Material and methods: This cross-sectional study included 187 participants, out of which 108 women with PCOS and 79 healthy women in the reproductive period as controls. Basal hormonal, metabolic, antropometric and clinical parameters for PCOS were obtained for all participants. For the first group of 60 women with PCOS and 31 controls, the levels of AMH were additionally measured. For the second group of 78 women with hyperinsulinemic PCOS and 80 normonisulinemic controls, basal and stimulated values of insulin and glycemia during oral glucose tolerance test (OGTT) were measured, and the value of HOMA-IR was determined for all. Results: The researched groups did not differ in the mean age (p=0,057), body mass index (p=0,057), waist circumference (p=0,026), hips circumference (p=0,691), systolic (p=0,716) and diastolic blood pressure (p=0,571). Significantly higher values of classical hormonal parameters for PCOS were observed in the PCOS group: total testosterone (p<0,001), free androgen index (p<0,001), LH (p<0,001), LH/FSH (p<0,001), DHEA-S (p<0,001) and androstenedione (p<0,001), with significantly lower values of SHBG (p<0,001). There was no statistically significant difference in the values of FSH (p=0,118), estrogen (p=0,050) and prolactin (0,952) between the groups. AMH values were significantly higher in the PCOS group (p<0,001), with highest values observed in the phenotype A, than phenotype B and phenotype D, and lowest values observed in phenotype C. The value of AUC-ROC for AMH was 0,961 (95% CI 0,926-0,996). At a cut-off value for AMH of 4,7 ng/ml we determined sensitivity of 90,0% and specificity of 90,3%, positive predictive value of 94,7% and negative predictive value of 82,4% for the method. Significant correlations between AMH and duration of menstrual cycle (p<0,001), LH (p=0,025), LH/FSH (p<0,001), total testosterone (p=0,034), and free androgen index (p=0,034) were observed, as well as negative correlelation of AMH with FSH (p=0,011) and basal insulin (p=0,034). In the metabolic parameters, there was significant difference between the values of both basal (p<0,001), stimulated insulin at 60’ (p<0,001) and stimulated insulin at 120’ during OGTT (p<0,001) between groups. Only the values of basal glycemia were comparable between the groups (p=0,119), while higher stimulated values of glycemia at 60’ (p<0,001) and at 120’ (p<0,001) were observed. HOMA-IR values were higher (p<0,001) in the hyperinsulinemic PCOS group in comparison to the HOMA-IR values in the normoinsulinemic group. AUC-ROC for HOMA-IR in determination of hyperinsulinemia in PCOS was 0,830 (95% CI 0,762-0,899). At a cut-off value for HOMA-IR of 2,5 we obtained sensitivity of 82,89% of the method for determining insulin resistance in PCOS, with specificity of 85,90%, positive predictive value of 85,14% and negative predictive value of 83,75%. We demonstrated positive correlation of HOMA-IR not only with basal insulin concentrations, but also with the stimulated insulin concentrations. Presence of dyslipidemia characteristic of insulin ressistence (low HDL and high triglycerides) was determined in the PCOS group. Conclusion: The high sensitivity and specificity obtained for AMH, as well as the association of AMH with hyperandrogenism, positions this method as an additional marker in diagnosing the severity of PCOS. The correlation of HOMA-IR not only with the basal but also with the stimulated insulin values opens possibilities for broader use of the method for assessing insulin resistance in women with PCOS in a clinical setting, with high sensistivity and specificity.